检索结果-内蒙古大学图书馆

Mamba in Speech: Towards an Alternative to Self-Attention

IEEE Transactions on Audio, Speech and Language Processing 2025年 33卷 1933-1948页

作者： Xiangyu Zhang Qiquan Zhang Hexin Liu Tianyi Xiao Xinyuan Qian Beena Ahmed Eliathamby Ambikairajah Haizhou Li Julien Epps School of Electrical Engineering and Telecommunications The University of New South Wales (UNSW) Sydney NSW Australia College of Computing and Data Science Nanyang Technological University Singapore School of Computer and Communication Engineering University of Science and Technology Beijing Beijing China Guangdong Provincial Key Laboratory of Big Data Computing The Chinese University of Hong Kong (Shenzhen) Shenzhen China Shenzhen Research Institute of Big Data Shenzhen China

Transformer and its derivatives have achieved success in diverse tasks across computer vision, natural language processing, and speech processing. To reduce the complexity of computations within the multi-head self-attention mechanism in Transformer, Selective State Space Models (i.e., Mamba) were proposed as an alternative. Mamba exhibited its effectiveness in natural language processing and computer vision tasks, but its superiority has rarely been investigated in speech signal processing. This paper explores solutions for applying Mamba to speech processing by discussing two typical speech processing tasks: speech recognition, which requires semantic and sequential information, and speech enhancement, which focuses primarily on sequential patterns. The experimental results confirm that bidirectional Mamba (BiMamba) consistently outperforms vanilla Mamba, highlighting the advantages of a bidirectional design for speech processing. Moreover, experiments demonstrate the effectiveness of BiMamba as an alternative to the self-attention module in the Transformer model and its derivates, particularly for the semantic-aware task. The crucial technologies for transferring Mamba to speech are then summarized in ablation studies and the discussion section, offering insights for extending this research to a broader scope of tasks.

关键词： Transformers Speech enhancement Speech recognition Convolution Computational modeling Natural language processing Training Random access memory Kernel Semantics

来源：评论

学校读者我要写书评

暂无评论

Personalized Federated Learning for Smart Healthcare Using SDN-Driven Edge Computing

Personalized Federated Learning for Smart Healthcare Using S...

引用

2025 IEEE SoutheastCon, SoutheastCon 2025

作者： Basu, Deborsi Chaudhary, Ritu Ghosh, Uttam Datta, Raja G. S. Sanyal School of Telecommunications Indian Institute of Technology Kharagpur India Indian Institute of Technology Dept. of Electronics and Electrical Communication Engineering Kharagpur India School of Applied Computational Sciences Dept. of Computer Science and Data Science Mmc TN United States

ISBN: (纸本)9798331504847

The advancement in distributed Federated Learning (FL) techniques is transforming next-generation edge computing, especially in e-Helathcare systems where data integrity and security are critical concerns. Collecting, processing and analyzing such substantial medical data is highly challenging due to their non-identical and versatile nature. Conventional FL approaches often show unsatisfactory and inaccurate results while operating on those data. Traditional FL methods also lack optimal dynamic management mechanisms for edge devices in ehealthcare, which is crucial to ensure that only authorized devices participate in handling any system failures and new registrations. To tackle such important aspects, in this work, we propose a personalized FL technique to handle abrupt label shifts among clients' data. The proposed approach further targets to achieve personalized models for local edge clients. A statistical batch normalization is proposed on customized edge servers, keeping the data specificity intact. To make the proposed system more secure and robust, a Software Defined Network-based (SDN-based) centralized server is used that allows only authorized and efficient edge clients to participate. The proposed comprehensive experiments on two healthcare datasets demonstrate better accuracy than traditional FL methods with enhanced security and privacy. © 2025 IEEE.

关键词： mHealth

来源：评论

学校读者我要写书评

暂无评论

Movable Antenna Enabled Near-Field Communications: Channel Modeling and Performance Optimization

引用

IEEE Transactions on Communications 2025年

作者： Zhu, Lipeng Ma, Wenyan Xiao, Zhenyu Zhang, Rui National University of Singapore Department of Electrical and Computer Engineering 117583 Singapore Beihang University School of Electronic and Information Engineering Beijing100191 China The Chinese University of Hong Kong School of Science and Engineering Shenzhen Research Institute of Big Data Shenzhen518172 China

Movable antenna (MA) technology offers promising potential to enhance wireless communication by allowing flexible antenna movement. To maximize spatial degrees of freedom (DoFs), larger movable regions are required, which may render the conventional far-field assumption for channels between transceivers invalid. In light of it, we investigate in this paper MA-enabled near-field communications, where a base station (BS) with multiple movable subarrays serves multiple users, each equipped with a fixed-position antenna (FPA). First, we extend the field response channel model for MA systems to the near-field propagation scenario. Next, we examine MA-aided multiuser communication systems under both digital and analog beamforming architectures. For digital beamforming, spatial division multiple access (SDMA) is utilized, where an upper bound on the minimum achievable rate across users is derived in closed form. A low-complexity algorithm based on zero-forcing (ZF) is then proposed to jointly optimize the antenna position vector (APV) and digital beamforming matrix (DBFM) to approach this bound. For analog beamforming, orthogonal frequency division multiple access (OFDMA) is employed, and an upper bound on the minimum achievable rate among users is also derived. An alternating optimization (AO) algorithm is proposed to iteratively optimize the APV, analog beamforming vector (ABFV), and power allocation until convergence. For both architectures, we further explore MA design strategies based on statistical channels, with the APV updated less frequently to reduce the antenna movement overhead. Simulation results demonstrate that our proposed algorithms achieve performance close to the derived bounds and also outperform the benchmark schemes using dense or sparse arrays with FPAs. 1972-2012 IEEE.

关键词： Near field communication

来源：评论

学校读者我要写书评

暂无评论

Analysis of converting -circuit into -circuit

引用

EPJ Quantum Technology 2025年第1期12卷 1-14页

作者： Luo, Qing-bin Ding, Lang Yang, Guo-wu Li, Xiao-yu College of Intelligent Systems Science and Engineering Hubei Minzu University Enshi China School of Information and Software Engineering University of Electronic Science and Technology of China Chengdu China Big data research Center & School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu China

A $\mathfrak{C^{*}}$ -circuit, which was proposed in Asiacypt 2022 by Huang and Sun (Advances in cryptology – ASIACRYPT 2022, pp. 614–644, 2022), can directly perform calculations with the existing quantum states, thereby reducing the use of quantum resources in quantum logic synthesis. We theoretically prove how to convert a $\mathfrak{C^{0}}$ -circuit into the corresponding $\mathfrak{C^{*}}$ -circuit through two lemmas and one theorem. The first lemma proves the interchangeability of CNOT gates and NOT gates by using the equivalence of quantum circuits. The second lemma proves that adding CNOT gates to the front of a quantum circuit whose initial states are all $|0\rangle $ s will not change the output states of the circuit. The theorem is used to describe what kind of $\mathfrak{C^{0}}$ -circuit can be transformed into $\mathfrak{C^{*}}$ -circuit, and the correctness of this transformation is proved. Our work will provide a theoretical basis for converting $\mathfrak{C^{0}}$ -circuit to $\mathfrak{C^{*}}$ -circuit. Then applying the theoretical analysis results to the multiplication over $\text{GF}(2^{8})$ , the constructed quantum circuit needs 27 Toffoli gates and 118 CNOT gates, which is 15 fewer Toffoli gates and 43 CNOT gates than the current best result. This shows that the method of constructing quantum circuits by using the conversion of $\mathfrak{C^{0}}$ -circuit to $\mathfrak{C^{*}}$ -circuit is very efficient.

关键词：

来源：评论

学校读者我要写书评

暂无评论

AuthUp: A Privacy-Preserving V2I Authentication Scheme for Vehicular Ad Hoc Networks

引用

IEEE Transactions on Vehicular Technology 2025年

作者： Li, Xincheng Yin, Xinchun Ning, Jianting Zhang, Tianwei Zhejiang Sci-Tech University School of Information Science and Engineering School of Cyber Science and Technology the Zhejiang Key Laboratory of Digital Fashion and Data Governance Hangzhou310018 China Jiangxi Vocational College of Science and Technology Nanchang330200 China Wuhan University School of Cyber Science and Engineering Wuhan430072 China City University of Macau Faculty of Data Science China College of Computer and Cyber Security Fuzhou350117 China Nanyang Technological University School of Computer Science and Engineering 639798 Singapore

Vehicular ad hoc networks (VANETs) play a crucial role in enhancing modern transportation systems, leading to the development of various authentication schemes to meet the stringent security and privacy requirements of vehicular communications. Many existing schemes rely on tamper- proof devices (TPDs) that store extensive secret parameters for identity authentication, making them susceptible to Sybil and side-channel attacks. This paper introduces a sophisticated privacy-preserving authentication protocol, referred to as Authentication and Update (AuthUp), specifically designed to solve the problem. A trusted authority (TA) issues one-time access tokens to vehicles during the registration process. After each vehicle-to-infrastructure (V2I) authentication process, cryptographic key material is updated without compromising privacy. Consequently, Sybil attacks are promptly identified and thwarted, as manipulated tokens fail authentication. Additionally, adversaries are prevented from accessing sensitive data stored in TPDs via side-channel attacks because of the dynamic updating of secret keys. A formal security proof validates the robustness of AuthUp, and extensive evaluations demonstrate its superior efficiency compared to alternative schemes, highlighting its suitability for VANET deployment. © 1967-2012 IEEE.

关键词： Side channel attack

来源：评论

学校读者我要写书评

暂无评论

AIGC-Driven Real-Time Interactive 4D Traffic Scene Generation in Vehicular Networks

引用

IEEE Network 2025年

作者： Li, Xiaolong Deng, Ruiting Wei, Jianhao Wu, Xin Chen, Jiayuan Yi, Changyan Cai, Jun Niyato, Dusit Shen, Xuemin Hunan University of Technology and Business School of Computer Science China Nanjing University of Aeronautics and Astronautics College of Computer Science and Technology China Concordia University Department of Electrical and Computer Engineering Montreal Canada Nanyang Technological University College of Computing and Data Science Singapore University of Waterloo Department of Electrical and Computer Engineering Canada

Real-time, interactive 4D traffic scene generation enables rapid digital twinning of traffic scenarios, improving management and decision-making in intelligent transportation systems. However, current text-to-video models, such as Sora, struggle to maintain the temporal coherence of traffic elements and interact with dynamic environments and users when generating 4D scenes. This article introduces a novel cloud-edge-terminal collaborative framework that leverages Artificial Intelligence-Generated Content (AIGC) in vehicular networks to tackle these challenges, ensuring long-term coherence and improved interactivity. The framework presents a comprehensive architecture for real-time interactive 4D scene generation, encompassing data collection, management, model pre-training, fine-tuning, and inference. We examine key design requirements and challenges, demonstrating that our microservice-based framework enables the system to generate and update 4D traffic scenes in real time, effectively responding to traffic data and user inputs. To the best of our knowledge, this is the first successful implementation of real-time, interactive 4D traffic scene generation. Performance evaluations show the superiority of our framework, powered by microservice-based code fine-tuning, over traditional frameworks. Finally, we discuss future research directions to enhance AIGC-driven 4D traffic scene generation. © 1986-2012 IEEE.

关键词： Decision making

来源：评论

学校读者我要写书评

暂无评论

Energy Consumption Optimization using Big data Analytics for Smart Grid Management 5

Energy Consumption Optimization using Big Data Analytics for...

引用

5th International Conference on Emerging Systems and Intelligent Computing, ESIC 2025

作者： Devi, K. Sumithra Contractor, Dipenkumar Deepa, V. Lakshmi, T.K. Santhikala, M. Mounika, Kamati Kcg College of Technology Department of Computer Science and Engineering Chennai India Government Engineering College Department of Computer Engineering Dahod India Sri Ramakrishna College of Arts and Science Department of Information Technology Tamil Nadu Coimbatore India Malla Reddy University Department of Computer Science and Engineering Telangana Hyderabad India Ramco Institute of Technology Department of Artificial Intelligence and Data Science Tamil Nadu Rajapalayam India Mb School of Pharmaceutical Sciences Mohan Babu University Department of Pharmacy Practice Andhra Pradesh Tirupathi India

ISBN: (纸本)9798331522100

The application of big data analytics to improve energy usage efficiency in smart grids has been prompted by the rising need for power management. Adding to the above, it poses a big data solution for demand prediction, load consolidation, and fault diagnosis. Establishing the new and state-of-the-art innovative LSTM network that was employed in the forecast yielded a higher demand forecast accuracy of 94.2% compared to the previous systems. The load balancing index achieved 92% with an accuracy of 93.7% in fault detection, thus minimizing maintenance cost by 14.7% while improving grid reliability. The proposed model was found to give better results than previous conventional machine learning techniques and even modern ones which are more complex, especially for an ever-changing grid environment. Overall, these findings show the promise of big data solutions for handling the challenges of the contemporary smart grid while improving its efficiency through energy optimization. Consequently, this work lays a framework for subsequent studies on the remaining applications of advanced analytics in energy management systems. © 2025 IEEE.

关键词： Smart power grids

来源：评论

学校读者我要写书评

暂无评论

Phenotype and Genotype Based Sample Aware Surrogate-Assisted Genetic Programming in Dynamic Flexible Job Shop Scheduling

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2025年

作者： Zhu, Luyao Zhang, Fangfang Zhu, Xiaodong Chen, Ke Zhang, Mengjie Zhengzhou University School of Electrical and Information Engineering Zhengzhou450001 China Victoria University of Wellington Centre for Data Science and Artificial Intelligence School of Engineering and Computer Science Wellington6140 New Zealand Longmen Laboratory Luoyang471000 China

Genetic programming (GP) has been widely applied to evolve scheduling heuristics for dynamic flexible job shop scheduling (DFJSS). However, the evaluation of GP individuals is computationally expensive, especially in large scale DFJSS scenarios. A k-nearest neighbor (KNN) based surrogate has been successfully used to reduce individual evaluation time for GP by predicting the fitness of an individual with the most similar sample in KNN. Particularly, the phenotypes of GP individuals have been utilised to generate samples for KNN-based surrogates with a precondition that the fitness of individuals with the same phenotype is the same or similar. However, their real fitness may differ greatly due to different input decision situations for fitness calculations in DFJSS. Thus, only considering phenotypes of GP individuals to extract samples could decrease the accuracy of KNN surrogates. This paper proposes a KNN-based surrogate assisted GP algorithm by considering both the phenotype and genotype of GP individuals to generate samples. Specifically, a genotypic characterisation based on terminal frequency is designed to measure the similarity of individual genotypes. The results show that with the same training time, the proposed algorithm can converge fast and achieve better scheduling heuristics than the state-of-the-art algorithms in most examined scenarios. With the same number of generations, the proposed algorithm can obtain comparable performance but only needs about one third training time of baseline GP. The effectiveness of the proposed algorithm is also verified from different aspects, e.g., relation between genotype correlation and fitness difference of individuals, and population diversity. © 2025 IEEE.

关键词： Genetic programming

来源：评论

学校读者我要写书评

暂无评论

Learning Simultaneous and Sequential Decisions in Multi-Agent Systems With Application to Traffic Signal Control

引用

IEEE Transactions on Intelligent Transportation Systems 2025年第6期26卷 8257-8267页

作者： Zhang, Haipeng Wang, Zhiwen Yu, Jilin Zhang, Ruonan Jiang, Caoqing Luo, Gongkun Wu, Weiwei Wang, Wanyuan Guangxi University of Science and Technology College of Computer Science and Technology Liuzhou545006 China Guangxi University of Science and Technology School of Electronic Engineering Liuzhou545006 China Guangxi Key Laboratory of Big Data in Finance and Economics Nanning545006 China Guangxi University of Finance and Economics School of Big Data and Artificial Intelligence Nanning545006 China Southeast University School of Computer Science and Engineering Nanjing210096 China

Efficient traffic signal control (TSC) has been one of the most useful ways for reducing urban road congestion. By modeling each intersection as an autonomous agent, multiagent reinforcement learning (MARL) shows remarkable performance in solving dynamic TSC. However, most of TSC methods based on MARL suffer from a non-stationarity problem since agents update their policies simultaneously. To resolve this issue, this paper considers multi-intersection TSC as a multi-agent sequential decision-making process with policy online learning. We utilize a sequential model such as Transformer architecture to learn the multi-agent joint policy. By carefully designing the advantage function of each agent, the monotonic improvement property can be guaranteed. Moreover, to fully exploit the advantages of both simultaneous and sequential MARL, we further propose a novel MARL network selection algorithm (MARL-NS) which selectively employs simultaneous MARL only at states that sequential MARL might fall into local optimum. Our theory proves that MARL-NS preserves cooperative MARL converge properties. Finally, we validate the proposed MARL-NS method on a unified TSC benchmark, LibSignal. Experimental results show that our method can outperform the baseline methods in network-level and arterial coordination. © 2000-2011 IEEE.

关键词： Traffic congestion

来源：评论

学校读者我要写书评

暂无评论

Joint State and Noise Covariance Estimation

arXiv

引用

arXiv 2025年

作者： Khosoussi, Kasra Shames, Iman School of Electrical Engineering and Computer Science The University of Queensland St LuciaQLD Australia School of Engineering The Australian National University ACT Canberra Australia CSIRO Robotics Data61 Australia

This paper tackles the problem of jointly estimating the noise covariance matrix alongside primary parameters (such as poses and points) from measurements corrupted by Gaussian noise. In such settings, the noise covariance matrix determines the weights assigned to individual measurements in the least squares problem. We show that the joint problem exhibits a convex structure and provide a full characterization of the optimal noise covariance estimate (with analytical solutions) within joint maximum a posteriori and likelihood frameworks and several variants. Leveraging this theoretical result, we propose two novel algorithms that jointly estimate the primary parameters and the noise covariance matrix. To validate our approach, we conduct extensive experiments across diverse scenarios and offer practical insights into their application in robotics and computer vision estimation problems with a particular focus on SLAM. Copyright © 2025, The Authors. All rights reserved.

关键词： SLAM robotics

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：